JJEM: Special Edition 2 (December 2024)
JJEM: Special Edition 2 (December 2024)
2024-12-08
Enhancing security in payment platform to prevent fraud.
Ruchitha H P, Dr. Raghavendra S P
To enhance one's consumer experiences and reduce loss of funds, monetary institutions must aggressively detect transaction risks. In this paper, we compare various machine learning methods for accurately and efficiently predicting the validity of financial transactions. The methods utilized in this work included MLP Repressor, Random Forest Classifier, Complement NB, MLP Classifier, Gaussian NB, Bernoulli NB, LGBM Classifier, Ada Boost Classifier, K Neighbors Classifier, Logistic Regression, Bagging Classifier, Decision Tree Classifier, and Deep Learning. The random forest classifier performed best with unbalanced datasets the accuracy is 97% the precession is 88% the recall rate is 89% and the score for f1 is 95% however using bagging classification performed best on a balanced dataset the accuracy precession recall and f1-score are all 95%. The dataset was collected from Kaggle repository. It consists of 6000 rows and 10 columns. Dataset name is online fraud
Fraud detection, Decision tree, Random Forest, Machine Learning, Anaconda prompt